A python script for training and education purpose simulating an Raman measurement (session) on a graphene flake on a WITec alpha300 RAS device. All the software related steps (from starting of the software, finding the sample position, alignment of the Raman signal, acquiring point spectra at sample positions with different layer numbers, till saving the data) are performed for a graphene flake with differnet layer numbers.
This small program can be used for training or education purpose, e.g. in the context of an online practical class. Therefore, no initial guidance is given in the program itself and no "back" button is added for the individual steps during the simulated session. The individual steps of the simulation, ~"slide by slide" are listed in the Raman_sim_instructions.txt file, visible after unzip of the images.zip directory.
Compatible with Windows, Linux and MAC operating systems
The program contains screenshots of licensed software. The screenshots and the instruction file for individual measurement steps Raman_sim_instructions.txt are password protected. Please send me a password request by e-mail to obtain the password which is needed to run the program: stefan.noisternig@univie.ac.at.
For guidance and more information on other measuring modes please also consider the official WITec Suite Help with included operation guides to each measuring mode (since WITec Suite 5.3).
an example is given for Linux Systems
-
download files: Raman_sim_v1.py, images.zip and images.z01 in a directory
-
combine zip archives images.zip & images.z01 into new zip archive:
- go to the directory with the files and open terminal:
zip -F images.zip --out images_new.zip
- go to the directory with the files and open terminal:
-
request password for zip archive: stefan.noisternig@univie.ac.at
-
extract the zip archive:
unzip images_new.zip
- the images folder and Raman_sim_v1.py must be in the same directory
- go to the directory and execute Raman_sim_v1.py in python 3
- e.g. in Linux:
python Raman_sim_v1.py
- e.g. in Linux:
The License is given for the python file Raman_sim_v1.py itself and does not inlcude the content in the compressed images folders.